INDEX

Explanations

numerical values and terms related to accuracy and significance in data

oai_token-act-pair · gpt-4o-mini Triggered by @bot

significant figures

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.32.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

脚注の使い方

-0.81

transQ

-0.81

AsUp

-0.74

 queſta

-0.71

 myſelf

-0.70

 MainAxisSize

-0.69

ſelves

-0.69

MethodManager

-0.69

 faſt

-0.69

 zuſammen

-0.68

POSITIVE LOGITS

 digits

1.02

 digit

0.92

 four

0.92

digits

0.89

six

0.83

Digits

0.81

digit

0.80

 three

0.80

 figures

0.75

 five

0.75

Activations Density 0.115%

numerical values and terms related to accuracy and significance in data

significant figures

No Comments

No Known Activations